Age | Commit message (Collapse) | Author |
|
|
|
scheduler in next major release
|
|
|
|
|
|
Update documentation of PCH builder
|
|
Notes added on inclusion of PCH source file and object file.
Note that PCH is not just MSVC-only, but also C++-only.
Signed-off-by: Mats Wichmann <mats@linux.com>
|
|
Add LIBLITERALPREFIX to support gcc -l:filename
|
|
|
|
Add StenGruener's --debug=sconscript
|
|
|
|
MSVC: adjust the compiler search and warning message when configuring the environment.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
* Beef up _stripixes unittests
* rename LIBLITERAL to LIBLITERALPREFIX
* Initialize in Platform code, takes an error if not set when
StringSubber.expand() is called with a string containing LIBLITERALPREFIX
* Fix some bad typing markup.
* Add docs
Signed-off-by: Mats Wichmann <mats@linux.com>
|
|
with a False default value, and add properly passing LIBLITERAL in link, dmd, ldc tools where _stripixes() is actually used in current SCons Code (#10)
|
|
This is done in a general way. using a construction var, although
at the moment only the GNU linker is known to handle things this way.
Had to do something funky, or it won't work when os.pathsep and
LIBLITRAL are the same, as they are on Linux (i.e. ':'). That's because
SCons.PathList.PathList is called and it does:
class _Pathlist:
def __init__(self, pathlist) -> None:
if SCons.Util.is_String(pathlist):
pathlist = pathlist.split(os.pathsep)
Splitting has the effect of turning ":libm.a" into ((0, ''), (0, 'libm.a')]
That is, the ':' is lost as part of the library specifier - so need to
try to avoid that.
Fixes #3951
Signed-off-by: Mats Wichmann <mats@linux.com>
|
|
|
|
Add support for Java 21
|
|
Released in Sept 2023 - LTS version.
Signed-off-by: Mats Wichmann <mats@linux.com>
|
|
Elminate http: references
|
|
Most remaining http: references are either changed to https: or removed
or changed in case they were also stale links, and a replacement was
findable.
Does not affect schema and stylesheet references, which can remain http:
Signed-off-by: Mats Wichmann <mats@linux.com>
|
|
MSCommon: add custom formatter for debug log records
|
|
MSVC: update msvc path normalization utility functions
|
|
SCons/Tool/MSCommon/MSVC/UtilTests.py.
|
|
Changes:
* Adjust os path separators in MSCommon/MSVC/Util.py
* Revise commend inside resolve_path method in in MSCommon/MSVC/Util.py
* Move Util import inside verify method in MSCommon/MSVC/Config.py (prevent import dependency loops).
|
|
|
|
Manually resolved conflicts:
* CHANGES.txt
|
|
|
|
Fix issues in new scons_subproc_run
|
|
Fixup one test framework test
|
|
Update assembly code in one test
|
|
Microsoft tools: change relative imports to top-level absolute imports
|
|
Changes:
* Replace process_path function with resolve_path and normalize_path functions in MSCommon/MSVC/Util.py.
* Replace process_path invocations with normalize_path invocations.
* Protect against inadvertent resolve/realpath for windows drive specifications in resolve_path and normalize_path.
* Additional options for normalize_path with defaults consistent with process_path.
|
|
an optional class name.
Changes:
- Add support for optional extra dictionary for the python logging.
- Allow passing 'classname' via the extra dictionary and creating the log records with '%(classname)s.%(funcName)s' for class methods.
- Add a method to populate and return the extra dictionary.
- Add a boolean constant indicating if debugging is enabled or not.
|
|
|
|
up the msvc environment.
Changes:
* The search for the msvc compiler executable (cl.exe) no longer inspects the OS system path in certain situations when setting up the msvc environment.
* When the msvc compiler executable is not found during setup of the msvc environment, the warning message issued takes into account whether or not a possibly erroneous compiler executable was already present in the scons environment path.
|
|
imports for the Microsoft tools.
Moving any of the Microsoft tools that used relative imports to the scons site tools folder would fail on import (i.e., the relative import paths become invalid when the tools are moved).
|
|
Avoids warnings, which fail the test.
Fixes #4238
Signed-off-by: Mats Wichmann <mats@linux.com>
|
|
Looking at wrong exceptions for possible rmdir failures
Signed-off-by: Mats Wichmann <mats@linux.com>
|
|
Further mod on msvc config cache locking
|
|
In the first iteration, we took a read lock in the read case.
A fail was observed where the reader took a json decode error,
and then tried to remove the cachefile in case it's really corrupt.
That failed as the file was "busy" - on Windows removing is
essentially writing the file - so take a write lock here too.
Signed-off-by: Mats Wichmann <mats@linux.com>
|
|
Signed-off-by: Mats Wichmann <mats@linux.com>
|
|
Documentation reworded (docstring/comment only, this is not a
public API).
Removed duplicated cleaning of execution env - previous iteration
both called the Util function and then did the same work manually.
Added some unit tests for the argument fiddling.
Signed-off-by: Mats Wichmann <mats@linux.com>
|
|
The logic to handle the legacy 'error' kwarg supported by the
predecssor _subproc function had some problems. Reworked.
A few minor lint tweaks also.
Signed-off-by: Mats Wichmann <mats@linux.com>
|
|
scons_subproc doc: remove keyword roles
|